Functions for generating song recommendations using Spotify data. Sample code for my Medium article: "How to build an amazing music recommendation algorithm."

recommend_songs.py

from collections import defaultdict
from scipy.spatial.distance import cdist
import difflib

number_cols = ['valence', 'year', 'acousticness', 'danceability', 'duration_ms', 'energy', 'explicit',
 'instrumentalness', 'key', 'liveness', 'loudness', 'mode', 'popularity', 'speechiness', 'tempo']

def get_song_data(song, spotify_data):
    
    """
    Gets the song data for a specific song. The song argument takes the form of a dictionary with 
    key-value pairs for the name and release year of the song.
    """
    
    try:
        song_data = spotify_data[(spotify_data['name'] == song['name']) 
                                & (spotify_data['year'] == song['year'])].iloc[0]
        return song_data
    
    except IndexError:
        return find_song(song['name'], song['year'])
        

def get_mean_vector(song_list, spotify_data):
  
    """
    Gets the mean vector for a list of songs.
    """
    
    song_vectors = []
    
    for song in song_list:
        song_data = get_song_data(song, spotify_data)
        if song_data is None:
            print('Warning: {} does not exist in Spotify or in database'.format(song['name']))
            continue
        song_vector = song_data[number_cols].values
        song_vectors.append(song_vector)  
    
    song_matrix = np.array(list(song_vectors))
    return np.mean(song_matrix, axis=0)

def flatten_dict_list(dict_list):
   
    """
    Utility function for flattening a list of dictionaries.
    """
    
    flattened_dict = defaultdict()
    for key in dict_list[0].keys():
        flattened_dict[key] = []
    
    for dictionary in dict_list:
        for key, value in dictionary.items():
            flattened_dict[key].append(value)
            
    return flattened_dict
        

def recommend_songs(song_list, spotify_data, n_songs=10):
  
    """
    Recommends songs based on a list of previous songs that a user has listened to.
    """
    
    metadata_cols = ['name', 'year', 'artists']
    song_dict = flatten_dict_list(song_list)
    
    song_center = get_mean_vector(song_list, spotify_data)
    scaler = song_cluster_pipeline.steps[0][1]
    scaled_data = scaler.transform(spotify_data[number_cols])
    scaled_song_center = scaler.transform(song_center.reshape(1, -1))
    distances = cdist(scaled_song_center, scaled_data, 'cosine')
    index = list(np.argsort(distances)[:, :n_songs][0])
    
    rec_songs = spotify_data.iloc[index]
    rec_songs = rec_songs[~rec_songs['name'].isin(song_dict['name'])]
    return rec_songs[metadata_cols].to_dict(orient='records')